EMBER: Autonomous Cognitive Behaviour from Learned Spiking Neural Network Dynamics in a Hybrid LLM Architecture

TL;DR

This paper introduces a biologically-inspired hybrid cognitive system integrating a large language model with a spiking neural network that autonomously triggers actions based on learned associations, demonstrating emergent reasoning.

Contribution

It presents a novel architecture that replaces traditional retrieval with a spiking neural network to autonomously initiate reasoning and actions in an LLM-based system.

Findings

The SNN achieves 82.2% discrimination retention across embedding dimensions.

The system autonomously initiated contact after 8 hours of idle time.

First SNN-triggered action occurred after only 7 conversational exchanges.

Abstract

We present (Experience-Modulated Biologically-inspired Emergent Reasoning), a hybrid cognitive architecture that reorganises the relationship between large language models (LLMs) and memory: rather than augmenting an LLM with retrieval tools, we place the LLM as a replaceable reasoning engine within a persistent, biologically-grounded associative substrate. The architecture centres on a 220,000-neuron spiking neural network (SNN) with spike-timing-dependent plasticity (STDP), four-layer hierarchical organisation (sensory/concept/category/meta-pattern), inhibitory E/I balance, and reward-modulated learning. Text embeddings are encoded into the SNN via a novel z-score standardised top-k population code that is dimension-independent by construction, achieving 82.2\% discrimination retention across embedding dimensionalities. We show that STDP lateral propagation during idle operation can trigger and shape LLM actions without external prompting or scripted triggers: the SNN determines when to act and what associations to surface, while the LLM selects the action type and generates content. In one instance, the system autonomously initiated contact with a user after learned person-topic associations fired laterally during an 8-hour idle period. From a clean start with zero learned weights, the first SNN-triggered action occurred after only 7 conversational exchanges (14 messages).